At the Administrative Data Research conference held in Cardiff in December 2019, Dennis Culhane presented work from the United States (US) examining an emerging ‘ageing crisis’ amongst homeless populations.

Using demographic methods applied to 20 years worth of data from homeless shelters in several major US cities, Culhane explored changes in the age composition of single homeless shelter users over time.

Culhane’s findings suggested that certain birth cohorts were at an increased risk of homelessness over the life-course. Of policy concern was the presence of a large group of single people currently in the 46 to 57 year old age range that were likely to become homeless, meaning that age related issues were likely to become ‘a substantial problem’ amongst single homeless populations.

What we did and what we found

Inspired by Culhane’s analysis, researchers at ADR Wales were keen to explore this in a UK context and, due to data availability, have undertaken analysis of aggregate homelessness data in Scotland. Data relating to homelessness in England, Wales, and Northern Ireland have been aggeragted by local authorities before being sent to their respective government departments, meaning that the further analysis of such data are limited to the data as it is aggregated. The age and household structure breakdowns were insufficient for the purposes of our analysis. However, Scottish Government collect individual level data, which enables re-analysis and querying of data.

By using data from Scotland, the research team have been able to benefit from the long running collection of homelessness data, which began in 2002/03. Furthermore, as the data are collected at an individual level, Scottish Government were able to easily provide new aggregate breakdowns by gender and whether the household was a single person or other household type; needed to replicate Culhane’s analysis.

Findings from the analysis tentatively suggest that, much like in the United States, there may be a birth cohort at higher risk of homelessness. Men from Generation X, born from the middle of the 1960s ending in the early 1980s, may be at a sustained risk of homelessness. For women, there was a far less pronouced evidence for an ageing cohort effect; instead, women seemed to have higher risk of homelessness at younger ages, regardless of their birth cohort. Further details can be found in the data insight produced by ADR Wales.

A tentative conclusion for policy and practice drawn from this analysis is the need for greater attention to the intersection of ageing, gender, and household structure in the design of services and policy. We may potentially see a homeless population in Scotland increasingly defined by older single men, with a seemingly persistent period of risk at younger ages for single women.

The research team plan to explore the use of Scottish individual level homelessness data in order to refine the analysis based on aggregated data. Working with individual level data will enable individuals to be linked over time, in order to examine whether the same individuals are ageing through the homeless system, or whether it is different individuals from the same birth cohort appearing at different points in time.

This blog was originally posted on the Scottish Centre for Administrative Data Research website.